Total Synthesis and Stereochemical Revision of Phacelocarpus 2-Pyrone A

Article abstract of DOI:10.1002/chem.201504089

The first total synthesis of phacelocarpus 2-pyrone A is reported. The original natural compound was tentatively assigned (by NMR spectroscopy) as containing two cis-alkenes and a trans-vinyl ether connected to a 2-pyrone ring motif. Our computational pre

Full text of DOI:10.1002/chem.201504089

DOI: 10.1002/chem.201504089
Communication
&
Natural Products
Total Synthesis and Stereochemical Revision of Phacelocarpus
-Pyrone A
2
Thomas O. Ronson, Michael J. Burns, Martin H. H. Voelkel, Kieren J. Evans, Jason M. Lynam,
[
a]
Richard J. K. Taylor,* and Ian J. S. Fairlamb*
Abstract: The first total synthesis of phacelocarpus 2-
pyrone A is reported. The original natural compound was
tentatively assigned (by NMR spectroscopy) as containing
two cis-alkenes and a trans-vinyl ether connected to a 2-
pyrone ring motif. Our computational predictions indicat-
ed that a cis-vinyl ether motif was equally feasible.
Attempts to prepare the trans-vinyl ether were met with
no success. The all cis-target compound was synthesised
in nine steps, employing key regio- and stereoselective re-
I
actions including Au-catalysed vinyl etherification, Wittig
alkenylation and end-game Stille macrocyclisation. Analy-
sis of the NMR data enabled identification and confirma-
tion of the correct structure of phacelocarpus 2-pyrone A,
containing a cis-vinyl ether. Our studies pave the way for
future development of methodologies to these structural-
ly distinct pyrone skipped-polyenyne natural products.
The phacelocarpus pyrones (e.g. 1–5, Figure 1) are a family of
remarkable pyrone-containing macrocycles isolated from
Phacelocarpus labillardieri, a marine red algae found abundant-
[1]
ly around the coasts of southern Australia. Despite limited
[2]
biological studies on these compounds, their distinctive ar-
rangement of skipped alkenes around a macrocyclic ring con-
taining an embedded 2- or 4-pyrone moiety makes them intri-
guing and challenging targets for synthetic chemists. Indeed,
no total synthesis of any member of this family had been com-
pleted since their isolation in the 1980s until earlier this year
when the first synthesis of brominated ether 3 was reported
Figure 1. Top: Various members of the phacelocarpus pyrone family includ-
ing phacelocarpus 2-pyrone A 1 (proposed structure); Bottom: Computed
lowest energy molecular structures of the E- and Z-vinyl ethers of phacelo-
carpus 2-pyrone A 1.
lenge. The structure of 1 was assigned on the basis of NMR
spectroscopic data and also by comparison to compounds E-2
and Z-2, and the stereochemistry at the enol ether bond tenta-
[
3]
by Fürstner’s group.
Of particular note amongst this unusual family of com-
pounds is phacelocarpus 2-pyrone A (1), the only member to
contain an unprecedented 2-pyronyl enol ether substructure
as part of the skipped unsaturated system. The incorporation
of these features into a 19-membered macrocycle makes the
construction of this compound a formidable synthetic chal-
[
1b]
tively assigned as the E-isomer. Since no total synthesis of
1 has thus far been reported, and following our recent success-
[
4]
ful completion of a model compound, we undertook to
develop a convergent strategy for the synthesis of 1 to
determine its authentic stereochemistry.
Molecular models revealed that the macrocycle containing
the E-vinyl ether was more strained than that containing the Z-
vinyl ether. DFT calculations (see the Supporting Information
for methodology) indicate that the energy difference between
the E/Z stereoisomers is small and that both should therefore
be viable.
[
a] T. O. Ronson, Dr. M. J. Burns, M. H. H. Voelkel, K. J. Evans, Dr. J. M. Lynam,
Prof. R. J. K. Taylor, Prof. I. J. S. Fairlamb
Department of Chemistry
University of York
York, YO10 5DD (UK)
E-mail: richard.taylor@york.ac.uk
ian.fairlamb@york.ac.uk
We anticipated that the major challenge in any synthetic
route aimed towards 1 would be the construction of the vinyl
ether linked to the electron-withdrawing 2-pyrone motif. The
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201504089.
Chem. Eur. J. 2015, 21, 18905 – 18909
18905
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
computational results detailed above indicated that it would
be desirable to prepare both E- and Z-vinyl ether isomers of 1,
and the route was mapped out with this in mind. In principle,
the vinyl ether can be formed in several ways, including a Pd-
catalysed Hartwig–Buchwald-type etherification reaction, as we
[
4a]
demonstrated in the synthesis of the arene mimetic of 1.
However, 4-hydroxy-2-pyrones possess limited nucleophilicity
(
pK ꢀ5) in these Pd-mediated reactions and do not form the
a
desired coupling products (see the Supporting Information for
I
[5]
details). Nolan’s Au-catalysed etherification methodology for
the reaction of substituted phenols with internal alkynes was
identified as a potentially efficient way to access this type of
motif, although at the point we initiated this work, no phenol
with a pK lower than 7.15 (p-NO C H OH) had been evaluated
a
2
6
4
as a coupling partner. This adventurous step would also raise
questions over the regio- and stereochemical outcome of the
reaction; we thus envisaged that a vinyl ether isomerization
reaction might become necessary later in the route.
Given the considerations above, the retrosynthetic analysis
of 1 was formulated as detailed in Scheme 1. Fragments A and
Scheme 2. Synthesis of compound 10 (fragment A). Real-time infrared spec-
troscopic analysis showing the formation of mono- and di-lithiated 2-pyrone
intermediates 8 and 9 (both are likely solvated species); conversions are nor-
malised against relative absorbance of each species (following IR bands at
À1
1
740 (7), 1698 (8) and 1634 (9) cm ). Reagents and conditions: a) TBDPSCl
(
(
(
1 equiv), imidazole (1 equiv), CH
2 equiv), THF, À788C, 30 min, then BF
1 equiv), 2 h, 91%; c) PPh (1.1 equiv), I
Cl
2
Cl
2
, RT, 26 h, 99%; b) 3 (2 equiv), nBuLi
·OEt (2 equiv), 15 min, then 4
(1.1 equiv), imidazole (2.2 equiv),
3
2
3
2
CH
2
2
, RT, 3 h, 88%; d) nBuLi (2.3 equiv), THF/HMPA (6:1), À788C, 30 min,
then 6 (1.5 equiv), 45 min, 71%. TBDPS=tert-butyldiphenylsilyl, Imid=imida-
zole, THF=tetrahydrofuran, HMPA=hexamethylphosphoramide.
Scheme 1. Retrosynthetic analysis to compound 1.
I
[8]
B would first be unified using the Au -catalysed etherification
ditions, was gathered to provide a deeper understanding of
step. We recognised that fragment A could be synthesised by
the process.
selective C-alkylation of the dilithium salt derived from 4-
Reaction monitoring with real-time infrared spectroscopic
[6]
TM
hydroxy-6-methyl-2-pyrone (7). The route then exploits the
double nucleophilic reactivity of (Z)-1-tributylstannyl-but-1-en-
analysis (using ReactIR IC10 with fixed Si probe) of 7 with
nBuLi in THF/HMPA at À788C, showed that the initial deproto-
nation of the hydroxyl group occurred within approximately
4 min, giving mono-lithium salt intermediate 8. The depletion
of 8 was concomitant with formation of dilithium species 9;
the reaction reached completion within 12 min. When the lith-
iation time in a preparative reaction was adjusted accordingly,
the desired product 10 could be obtained in 71% yield, with
no formation of the dialkylation side product 10’ observed.
With the synthesis of 10 completed, its utility in the Au-cata-
lysed etherification reaction of alkyne 13 was examined
(Scheme 3). The Au catalyst [{(IPr)Au} (m-OH)][BF ] is unusual,
[4a,7]
4
-triphenylphosphonium bromide (fragment C),
allowing
Wittig and Stille cross-coupling reactions to be assessed
sequentially as the closing steps in the synthesis. Obtaining
good regio- and stereochemical control would be crucial in
both of these intricate final steps.
The forward synthesis of 1 begins with alkyne 6, which can
be accessed in excellent yield from commercially available but-
[
4a]
3
6
9
-yn-1-ol 2 (Scheme 2). We found that the reaction of iodide
[6]
with the dilithium salt derived from 7, that is, intermediate
, was blighted by lower than expected yields of the desired
2
4
product 10, while formation of the unwanted C3-alkylation
side product 10’ was also observed. Given these issues, qualita-
tive kinetic information about the formation of mono-lithium
species 8 and dilithium species 9, under working reaction con-
existing in equilibrium with the Lewis acidic and Brønsted
I
[5]
basic monomeric Au species (11 and 12). Complexation of
I
Au to alkyne 13 is envisaged to form p-complex I. Here it can
be seen that two regioisomers can result by addition to either
Chem. Eur. J. 2015, 21, 18905 – 18909
www.chemeurj.org
18906
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Scheme 3. Possible reaction pathways of pyrone 7 with alkyne 13.
the a or b carbon centres. The stereochemistry in previously
reported reactions of internal alkynes is most commonly Z, but
we envisaged that a subsequent isomerisation at a late stage
might be feasible. Given that the construction of a trisubstitut-
ed vinyl ether containing a 2-pyrone by other methods had
previously proven inefficient or impossible (see the Supporting
[9]
[5]
Information for details), the Au-methodology was deemed
the most promising potential route.
We were pleased to establish that the reaction between 10
and 13 (5 equiv), catalysed by 1 mol% of [{(IPr)Au} (m-OH)]
2
[
BF ], gave 14 with excellent regioselectivity (b:a=10:1) and
4
complete Z-stereoselectivity in a yield of 76%. Moreover, intra-
or intermolecular addition of the 4-hydroxyl group to the inter-
nal alkyne present within 10 was successfully avoided by using
alkyne 13 in excess (Scheme 4). At this stage a number of dif-
Scheme 5. End-game synthesis of Z,Z,Z-1. Reagents and conditions: a) TBAF
(
1.1 equiv), THF, RT, 1.5 h, 89%; b) DMP (1.4 equiv), CH
c) 19 (2 equiv), NaHMDS (1.9 equiv), THF, À78 to 08C, then 16, À78 to RT,
h, 14% over two steps; d) 21 (10 mol%), LiCl (10 equiv), DMF, 358C, 18 h,
2 2
Cl , 08C to RT, 1 h;
1
Scheme 4. Au-catalysed reaction of 10 with 13 to give 12.
20%. TBAF=tetrabutylammonium fluoride, DMP=Dess–Martin periodinane,
NaHMDS=sodium hexamethyldisilazide, DMF=N,N-dimethylformamide,
PCC=pyridinium chlorochromate.
ferent isomerisation conditions were screened in an attempt to
obtain the E-enol ether from compound 14, but no isomerisa-
tion could be detected under any of the conditions screened
side reaction. This process could be catalysed by residual gold
remaining from the prior vinyl etherification step, which ac-
cording to literature reports can occur with as little as 10 ppm
(
see the Supporting Information for details). We thus resolved
to proceed with the total synthesis of Z-1 from compound 14,
which has a vinyl ether possessing Z-stereochemistry.
[
11]
Au. We therefore carried out the oxidation reaction with
DMP using starting material that had been treated with a thio-
urea-based resin (Quadrapure-TU) specifically designed to scav-
enge trace metal atoms; in this way, alcohol 15 was cleanly
oxidised to aldehyde 16. The latter compound is rather prone
to degradation, and thus 16 was used immediately in the
subsequent reaction.
The end-game synthesis of Z-1 is depicted in Scheme 5.
Alcohol 15 was revealed by silyl deprotection of 14 using
standard TBAF conditions, but initial attempts to oxidise the
primary alcohol 15 to aldehyde 16 under Dess–Martin condi-
[
10]
tions were not straightforward. The high polarity of the 2-
pyrone 16 meant that it was difficult to separate the desired
aldehyde cleanly from excess DMP and DMP-derived byprod-
ucts. Noted also was decomposition and side-product forma-
tion (e.g., furan 18 was isolated in 17% yield). Other oxidants,
for example, PDC and TPAP, led to the decomposition of 15,
whereas the use of PCC led to formation of 1,4-dicarbonyl
compound 17, in 57% yield.
The Wittig reaction with stannane–phosphonium salt 19 was
hampered by the particular sensitivity of aldehyde 16. At low
temperature (À788C), no reaction occurs between the ylide
formed from 19 and the aldehyde 16; upon warming, product
formation is observed but with accompanying decomposition.
Nevertheless, compound 20 could be prepared in 14% yield
over two steps, directly from compound 15. A detailed study
using NMR spectroscopic analysis of the Wittig reaction under
Dicarbonyl compound 17 is an obvious precursor to furan
18, thus pointing to unwanted alkyne hydration as a major
Chem. Eur. J. 2015, 21, 18905 – 18909
www.chemeurj.org
18907
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
working conditions showed that ylide oxidation was the
primary side reaction. Other attempts to optimize this reaction
were met with limited success.
framework. The stereochemistry of the vinyl ether linked to the
2-pyrone has been revised, by comparison of the reported
data of the natural compound with our data for the synthetic
natural product. We continue to work toward the synthesis of
other members of the 2-pyrone family of macrocycles, in
addition to expanding the Au-catalysed vinyl ether formation
reactions with 2-pyrones and related derivatives.
II
The Pd precatalyst, AsCat (21), which contains arsine, bro-
mide and succinimidyl groups, has previously been developed
[
12]
by our groups. It is a particularly active precatalyst for mild
Stille cross-coupling reactions, and was also applied to com-
[4a]
plete the synthesis of the arene mimetic of 1. The final Stille
macrocyclisation reaction of 20, mediated by AsCat (21), gave
Z,Z,Z-1 in 20% yield. All of the available NMR spectroscopic
data confirmed that the stereochemistry throughout the
molecule was all-cis (see the Supporting Information for
complete analysis and deductions).
Experimental Section
Experimental data for 1: R 0.58 (EtOAc/petrol, 1:1, v/v); IR (thin
f
À1
film, cm ) n
2925 s, 2854 m, 1733 s, 1645 m, 1567 m, 1462w,
max
1
1
6
2
4
417w, 1223 m, 1131w, 821w, 702w; H NMR (700 MHz, CDCl ) d
3
With spectroscopic data for synthetic Z,Z,Z-1 in hand, and
given that the DFT calculations had indicated that both Z- and
E-vinyl ethers are viable, we compared our data with the re-
.03 (d, J=2.2 Hz, 1H, H-22), 5.54–5.47 (m, 1H, H-10), 5.42 (d, J=
.2 Hz, 1H, H-21), 5.41–5.34 (m, 3H, H-6, 7, 9), 5.16 (t, J=7.3 Hz, H-
), 2.87 (d, J 7.5 Hz, 2H, H-11), 2.81 (t, J=7.5 Hz, 2H, H-8), 2.66 (t,
1
13
ported H and C NMR data for 1 (assigned with an E-vinyl
J=7.3 Hz, 2H, H-5), 2.51 (t, J=6.9 Hz, 2H, H-17), 2.23–2.16 (m, 4H,
H-14, 1’), 1.81 (app p, J=6.9 Hz, 2H, H-16), 1.62–1.53 (m, 2H, H-15),
[
1b]
ether stereochemistry).
To our delight, the data matched
1
3
1
3
1
3
3
.07 (t, J=7.4 Hz, 3H, H-2’); C NMR (see Table 1); MS (ESI) m/z (%)
very closely; the similarity in the C NMR data is particularly
striking (Table 1) in that all the carbon signals match within
+
61 ([M+Na] , 100); HRMS (ESI): m/z calcd for C H NaO :
61.1761 [M+Na] ; found: 361.1774.
22
26
3
+
13
Table 1. Comparison of reported C NMR spectroscopic data for 1 with
the data for the synthetic material.
Acknowledgements
Position
d (1, natural)
d (1, synthetic)
[ppm]
Dd
We thank Prof. S. P. Nolan for the kind donation of
{Au(NHC)} (m-OH)][BF ]. We are grateful to Dr. Natalie Fey (Uni-
[
a]
[b]
[ppm]
[ppm]
[
2
4
1
3
4
5
6
7
8
9
165.1
151.0
114.2
23.8
126.6
128.4
25.3
130.5
124.4
16.9
165.1
151.1
114.3
23.8
126.7
128.5
25.3
130.6
124.4
16.9
0
versity of Bristol) for advice and guidance concerning the con-
formational calculations detailed in Fig. 1. EPSRC (EP/J500598/
1, EP/D078776/1) and University of York are thanked for fund-
ing this work. I.J.S.F. would like to thank the Royal Society for
funding (University Research Fellowship). We are grateful to Dr.
Matthew Cliff (University of Manchester, which is part of the
N8 Universities) for the 800 MHz NMR experiments.
+0.1
+0.1
0
+0.1
+0.1
0
+0.1
0
0
1
0
11
1
1
1
1
1
1
1
2
2
2
1
2
2
3
4
5
6
7
8
0
1
2
’
78.9
79.3
18.1
27.3
25.0
32.3
166.9
169.0
89.6
79.0
79.4
18.1
27.3
25.0
32.4
166.9
169.0
89.6
+0.1
+0.1
0
0
Keywords: dienes
palladium · total synthesis
·
macrocycles
·
natural products
·
0
+0.1
[1] a) R. Kazlauskas, P. T. Murphy, R. J. Wells, A. J. Blackman, Aust. J. Chem.
1982, 35, 113–120; b) J. Shin, V. J. Paul, W. Fenical, Tetrahedron Lett.
1986, 27, 5189–5192; c) A. J. Blackman, J. B. Bremner, A. M. C. Paano,
J. H. Skerratt, M. L. Swann, Aust. J. Chem. 1990, 43, 1133–1136; d) L.
Murray, G. Currie, R. J. Capon, Aust. J. Chem. 1995, 48, 1485–1489.
[2] a) A. M. S. Mayer, V. J. Paul, W. Fenical, J. N. Norris, M. S. De Carvalho,
R. S. Jacobs, Hydrobiologia 1993, 260–261, 521–529; b) K. Sakata, Y.
Iwase, K. Kato, K. Ina, Y. Machiguchi, Nippon Suisan Gakkaishi 1991, 57,
0
0
0
0
0
0
99.0
25.5
11.0
99.0
25.5
11.0
’
[
a] From ref. [1b] (50 MHz); reassigned. [b] Referenced to CDCl
7.0 ppm (200 MHz). For the compound numbering system, see the Sup-
porting Information.
3
=
2
61–265.
3] L. Hoffmeister, T. Fukuda, G. Pototschnig, A. Fürstner, Chem. Eur. J. 2015,
1, 4529–4533; an elegant synthesis of a related natural product, neu-
7
[
2
rymenolide A, has been reported too, see: W. Chaładaj, M. Corbet, A.
Fürstner, Angew. Chem. Int. Ed. 2012, 51, 6929–6933; Angew. Chem.
2
012, 124, 7035–7039.
error the reported data for 1. The stereochemistry of phacelo-
carpus 2-pyrone A, 1, is therefore confirmed as all-cis. The
observation is supported by the computational study, which
indicates that the macrocycle with a Z-vinyl ether, relative to
the E-vinyl ether, is a feasible structure.
[
4] a) T. O. Ronson, M. H. H. Voelkel, R. J. K. Taylor, I. J. S. Fairlamb, Chem.
Commun. 2015, 51, 8034–8036. For an earlier model study, see: b) D.
Song, G. Blond, A. Fürstner, Tetrahedron 2003, 59, 6899–6904.
[5] Y. Oonishi, A. Gómez-Suµrez, A. R. Martin, S. P. Nolan, Angew. Chem. Int.
Ed. 2013, 52, 9767–9771; Angew. Chem. 2013, 125, 9949–9953.
6] X. Zhang, M. McLaughlin, R. L. P. MuÇoz, R. P. Hsung, J. Wang, J. Swidor-
ski, Synthesis 2007, 749–753.
7] E. J. Corey, M. d’Alarcao, K. S. Kyler, Tetrahedron Lett. 1985, 26, 3919–
3922.
[
In conclusion, we have synthesised phacelocarpus 2-pyrone
A, 1, and in the process have been able to confirm the
structural connectivity within this remarkable macrocyclic
[
Chem. Eur. J. 2015, 21, 18905 – 18909
www.chemeurj.org
18908
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[
[
8] For use of ReactIR in studying lithation chemistry, see: N. S. Sheikh, D.
Leonori, G. Barker, J. D. Firth, K. R. Campos, A. J. H. M. Meijer, P. A.
O’Brien, I. Coldham, J. Am. Chem. Soc. 2012, 134, 5300–5308.
9] a) T. O. Ronson, Ph. D. thesis, University of York (U.K.), 2015; b) M. J.
Burns, Ph. D. thesis, University of York (U.K.), 2010, available through
http://eprints.whiterose.ac.uk/; c) M. J. Burns, T. O. Ronson, R. J. K. Taylor,
I. J. S. Fairlamb, Beilstein J. Org. Chem. 2014, 10, 1159–1165.
[11] N. Marion, R. S. Ramón, S. P. Nolan, J. Am. Chem. Soc. 2009, 131, 448–
449.
[12] T. O. Ronson, J. R. Carney, A. C. Whitwood, R. J. K. Taylor, I. J. S. Fairlamb,
Chem. Commun. 2015, 51, 3466–3469.
Received: October 12, 2015
[
10] L. Wavrin, J. Viala, Synthesis 2002, 326–330.
Published online on November 16, 2015
Chem. Eur. J. 2015, 21, 18905 – 18909
www.chemeurj.org
18909
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim